Pipecat

Pipecat is an open-source Python framework that helps developers build real-time voice and multimodal AI agents by orchestrating speech, language, media, and application services. It provides composable pipelines for speech-to-text, large language models, text-to-speech, audio and video transport, turn-taking, and interruption handling, with integrations across multiple providers. Developers use it to create assistants, meeting tools, customer-support agents, and other conversational interfaces, including multi-agent workflows. The framework is maintained as a vendor-neutral open-source project with community contributions; teams can run it on their own infrastructure or use associated hosted and enterprise deployment offerings. Its users are software developers and organizations building production voice and multimodal applications.